Traffic sign detection based on HOG and PHOG using binary SVM and k-NN

Sugiharto, Aris; Harjoko, Agus

doi:10.1109/icitacee.2016.7892463

Cited by 15 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also conduct comparative experiments to demonstrate the effectiveness of our approach. Since no other algorithms that we are aware of address crater detection on natural terrains, we compare our framework with state-of-the-art object-recognition methods, HOG +SVM [25] and Convolutional Neural Network (CNN) [38][39][40]. In order to elicit a fair comparison, the standard HOG+SVM approach and CNN extracted feature with SVM (CNN +SVM) approach are applied to the satellite image in the same sliding window manner as our first stage framework.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…particularly well-suited method, given that crater detection is a target-specific learning task with a relatively small number of samples available. Since bomb craters generally follow isotropic patterns, the framework considers both shapes and appearances features, including circular shapes [22][23][24], contours [25], morphological features [26], and gradients [27]. When building these custom features, this framework accommodates the variation of shapes and surrounding objects since some craters have eroded or have been planted in the fifty years following the bombing.…”

Section: Competing Interestsmentioning

confidence: 99%

See 1 more Smart Citation

Crater detection from commercial satellite imagery to estimate unexploded ordnance in Cambodian agricultural land

et al. 2020

View full text Add to dashboard Cite

Unexploded ordnance (UXO) pose a significant threat to post-conflict communities, and current efforts to locate bombs rely on time-intensive and dangerous in-person enumeration. Very high resolution (VHR) sub-meter satellite images may offer a low-cost and high-efficiency approach to automatically detect craters and estimate UXO density. Machine-learning methods from the meteor crater literature are ill-suited to find bomb craters, which are smaller than meteor craters and have high appearance variation, particularly in spectral reflectance and shape, due to the complex terrain environment. A two-stage learning-based framework is created to address these challenges. First, a simple and loose statistical classifier based on histogram of oriented gradient (HOG) and spectral information is used for a first pass of crater recognition. In a second stage, a patch-dependent novel spatial feature is developed through dynamic mean-shift segmentation and SIFT descriptors. We apply the model to a multispectral WorldView-2 image of a Cambodian village, which was heavily bombed during the Vietnam War. The proposed method increased true bomb crater detection by over 160 percent. Comparative analysis demonstrates that our method significantly outperforms typical object-recognition algorithms and can be used for wide-area bomb crater detection. Our model, combined with declassified records and demining reports, suggests that 44 to 50 percent of the bombs in the vicinity of this particular Cambodian village may remain unexploded.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Competing Interestsmentioning

confidence: 99%

Crater detection from commercial satellite imagery to estimate unexploded ordnance in Cambodian agricultural land

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Traffic sign detection can be broadly divided into two categories [1][2][3][4][5][6]. One is the traditional method based on manual features [1][2][3][4][5], and the other is the deep learning algorithm based on CNN (Convolutional Neural Network) [6]. Traditional traffic signs are mainly detected based on the appearance characteristics of some traffic signs.…”

Section: Introductionmentioning

confidence: 99%

“…It reflects the matching degree between j-th target an i-th trajectory. If the Mahalanobis distance between the bounding box of the target in the current frame and the previous trajectory prediction observation is less than 9.4877, expressed as t (1) , the corresponding target and trajectory are related, expressed as b i,j = 1[d (1) (i, j) ≤ t (1) ]…”

mentioning

confidence: 99%

A New Real-Time Detection and Tracking Method in Videos for Small Target Traffic Signs

Song

Huang

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

It is a challenging task for self-driving vehicles in Real-World traffic scenarios to find a trade-off between the real-time performance and the high accuracy of the detection, recognition, and tracking in videos. This issue is addressed in this paper with an improved YOLOv3 (You Only Look Once) and a multi-object tracking algorithm (Deep-Sort). First, data augmentation is employed for small sample traffic signs to address the problem of an extremely unbalanced distribution of different samples in the dataset. Second, a new architecture of YOLOv3 is proposed to make it more suitable for detecting small targets. The detailed method is (1) removing the output feature map corresponding to the 32-times subsampling of the input image in the original YOLOv3 structure to reduce its computational costs and improve its real-time performances; (2) adding an output feature map of 4-times subsampling to improve its detection capability for the small traffic signs; (3) Deep-Sort is integrated into the detection method to improve the precision and robustness of multi-object detection, and the tracking ability in videos. Finally, our method demonstrated better detection capabilities, with respect to state-of-the-art approaches, which precision, recall and mAP is 91%, 90%, and 84.76% respectively.

show abstract

“…In [24], a supervised learning model called SVM-HOG is used for a car detection methodology in an outdoor environment. In [25], the authors compare k-Nearest Neighbor (k-NN) and SVM; SVM-HOG is better than k-NN-HOG. With SVM-HOG, we can recognize symbol characters on the screen of the smartphones and ignore the effects of lightness and camera angle.…”

Section: Introductionmentioning

confidence: 99%

Robotic-Based Touch Panel Test System Using Pattern Recognition Methods

Juang

2020

Applied Sciences

View full text Add to dashboard Cite

In this study, pattern recognition methods are applied to a five-degrees-of-freedom robot arm that can key in words on a touch screen for an automatic smartphone test. The proposed system can recognize Chinese characters and Mandarin phonetic symbols. The mechanical arm is able to perform corresponding movements and edit words on the screen. Pattern matching is based on the Red-Green-Blue (RGB) RGB color space and is transformed to binary images for higher correct rate and geometric matching. A web camera is utilized to capture patterns on the tested smartphone screen. The proposed control scheme uses a support vector machine with a histogram of oriented gradient classifier to recognize Chinese Mandarin phonetic symbols and provide correct coordinates during the control process. The control scheme also calculates joint angles of the robot arm during the movement using the Denavit–Hartenberg parameters (D-H) model and fuzzy logic system. Fuzzy theory is applied to use the position error between the robot arm and target location then resend the command to adjust the arm’s position. From the experiments, the proposed control scheme can control the robot to press desired buttons on the tested smartphone. For Chinese Mandarin phonetic symbols, recognition accuracy of the test system can reach 90 percent.

show abstract

Traffic sign detection based on HOG and PHOG using binary SVM and k-NN

Cited by 15 publications

References 10 publications

Crater detection from commercial satellite imagery to estimate unexploded ordnance in Cambodian agricultural land

Crater detection from commercial satellite imagery to estimate unexploded ordnance in Cambodian agricultural land

A New Real-Time Detection and Tracking Method in Videos for Small Target Traffic Signs

Robotic-Based Touch Panel Test System Using Pattern Recognition Methods

Contact Info

Product

Resources

About